The investigation first focussed on the failure of the input drive coupling between the engine and the main-rotor transmission. Close examination of the coupling indicated that it had been compressed and pulled apart. The extent of transmission and/or engine movement to facilitate this disconnect was not possible under any flight conditions. Furthermore, damage from the hard landing showed that the main rotor transmission rocked fore and aft, deformed its mounts, and allowed the transmission to tilt forward enough to pull the coupling apart. It was, therefore, concluded that all of the damage to the helicopter, excluding the engine, was the result of the hard landing and rollover. The following analysis focuses on the engine. The vibrations experienced when the engine was run in the test facility were extreme enough that they would have been detected by the pilot had they been present before the occurrence. Also, because the engine did not appear to suffer any damage from the hard landing, it was concluded that the imbalance occurred at about the time the smoke came into the cabin. This imbalance was determined to have been caused by the separation of parts of the GPturbine blades. There were no periodic power checks carried out, or required, by the operator to detect a power loss, nor maintenance inspections required by the manufacturer or the operator for sulfidation corrosion. It is suggested that either power checks or maintenance inspections would have led to the discovery of the corroded turbine components. The identified sulfidation weakened the GPturbine blades, fatigue cracks occurred on many of the blades, and one of the blades separated. Parts of other GP turbine blades broke off and engine power was lost. The loss of the blade and other blade sections caused an imbalance, which then likely resulted in damage to the turbine seals. It is likely that the seal damage allowed the release of oil onto hot engine areas, resulting in the soot in the turbine at the joint between the turbine nozzle and the number1 GPturbine and the smoke in the cabin. Another scenario is that the smoke came from the burnt boot for the input drive coupling. This scenario, however, is not supported by the information gathered during the investigation, which clearly indicates that the coupling failed during the hard landing. The pilot reacted immediately to the smoke by lowering the collective. The helicopter was operating in a hover, about 150feet above the ground at a relatively high density altitude, and at a weight near the maximum allowable. Regardless of the reason for the smoke or when power was lost, a successful landing with reduced power was unlikely. As the pilot tried to arrest the helicopter's descent, rotor rpm decayed, and not enough thrust was available to cushion the landing. When the landing gear skid tubes and forward cross tube broke, the cross tubes rotated aft and up. The after-market step that was attached to the cross tubes came up against the pilot door and the rotated cross tubes caused the step to be jammed against the door, preventing it from opening normally, and there was no emergency release system for the doors or windows. Because the helicopter was on its side, the other door was not usable, and the pilot and front passenger were trapped inside the helicopter. The following TSB Engineering Laboratory report was completed: LP 037/04 - Gas Producer Wheel, Bell Helicopter 206B, C-GPOS This report is available from the Transportation Safety Board of Canada upon request.Analysis The investigation first focussed on the failure of the input drive coupling between the engine and the main-rotor transmission. Close examination of the coupling indicated that it had been compressed and pulled apart. The extent of transmission and/or engine movement to facilitate this disconnect was not possible under any flight conditions. Furthermore, damage from the hard landing showed that the main rotor transmission rocked fore and aft, deformed its mounts, and allowed the transmission to tilt forward enough to pull the coupling apart. It was, therefore, concluded that all of the damage to the helicopter, excluding the engine, was the result of the hard landing and rollover. The following analysis focuses on the engine. The vibrations experienced when the engine was run in the test facility were extreme enough that they would have been detected by the pilot had they been present before the occurrence. Also, because the engine did not appear to suffer any damage from the hard landing, it was concluded that the imbalance occurred at about the time the smoke came into the cabin. This imbalance was determined to have been caused by the separation of parts of the GPturbine blades. There were no periodic power checks carried out, or required, by the operator to detect a power loss, nor maintenance inspections required by the manufacturer or the operator for sulfidation corrosion. It is suggested that either power checks or maintenance inspections would have led to the discovery of the corroded turbine components. The identified sulfidation weakened the GPturbine blades, fatigue cracks occurred on many of the blades, and one of the blades separated. Parts of other GP turbine blades broke off and engine power was lost. The loss of the blade and other blade sections caused an imbalance, which then likely resulted in damage to the turbine seals. It is likely that the seal damage allowed the release of oil onto hot engine areas, resulting in the soot in the turbine at the joint between the turbine nozzle and the number1 GPturbine and the smoke in the cabin. Another scenario is that the smoke came from the burnt boot for the input drive coupling. This scenario, however, is not supported by the information gathered during the investigation, which clearly indicates that the coupling failed during the hard landing. The pilot reacted immediately to the smoke by lowering the collective. The helicopter was operating in a hover, about 150feet above the ground at a relatively high density altitude, and at a weight near the maximum allowable. Regardless of the reason for the smoke or when power was lost, a successful landing with reduced power was unlikely. As the pilot tried to arrest the helicopter's descent, rotor rpm decayed, and not enough thrust was available to cushion the landing. When the landing gear skid tubes and forward cross tube broke, the cross tubes rotated aft and up. The after-market step that was attached to the cross tubes came up against the pilot door and the rotated cross tubes caused the step to be jammed against the door, preventing it from opening normally, and there was no emergency release system for the doors or windows. Because the helicopter was on its side, the other door was not usable, and the pilot and front passenger were trapped inside the helicopter. The following TSB Engineering Laboratory report was completed: LP 037/04 - Gas Producer Wheel, Bell Helicopter 206B, C-GPOS This report is available from the Transportation Safety Board of Canada upon request. Periodic power checks or inspections for sulfidation corrosion, which most likely would also have detected a power loss and the corroded turbine components, were not undertaken, nor were they required. A blade in the number1 GPturbine failed as a result of fatigue, which resulted from sulfidation corrosion; the engine lost power as a result of the blade failure. The helicopter was operating in a high hover when the engine lost power. The ensuing forced landing was hard, the landing gear broke, and the helicopter rolled over.Findings as to Causes and Contributing Factors Periodic power checks or inspections for sulfidation corrosion, which most likely would also have detected a power loss and the corroded turbine components, were not undertaken, nor were they required. A blade in the number1 GPturbine failed as a result of fatigue, which resulted from sulfidation corrosion; the engine lost power as a result of the blade failure. The helicopter was operating in a high hover when the engine lost power. The ensuing forced landing was hard, the landing gear broke, and the helicopter rolled over. The after-market step that was attached to the landing gear cross tubes blocked the pilot's door and trapped the pilot and passenger in the wreckage. Bell 206 series helicopters are not normally equipped with doors or windows that have emergency release systems.Findings as to Risk The after-market step that was attached to the landing gear cross tubes blocked the pilot's door and trapped the pilot and passenger in the wreckage. Bell 206 series helicopters are not normally equipped with doors or windows that have emergency release systems. On 18 July 2003, the TSB sent a Safety Information Letter to TC, with copies to Bell Helicopter Textron; Aeronautical Accessories, Inc.; British Columbia Forest Service; and Cariboo Chilcotin Helicopters Ltd. The letter described the conditions that resulted in the pilot's door being blocked by the after-market step. This information was provided to TC, to take whatever action it deemed necessary. On 03 October 2003, the TSB sent a Safety Advisory (A030018) to TC, with copies to Rolls-Royce, Bell Helicopter Textron, and Cariboo Chilcotin Helicopters Ltd., drawing attention to the predominant cause for the engine power loss - sulfidation corrosion. Because periodic inspections are not required, the risk of this corrosion going undetected in other aircraft powered by Rolls-Royce 250engines remains. The letter stated that TC may wish to consider whether periodic inspections should be carried out to help detect damage caused by sulfidation corrosion. TC responded to the Safety Advisory on 05February2004, suggesting that in the absence of data substantiating the need for scheduled washings of the compressor (and turbine) blades, the introduction of such a schedule is deemed unnecessary. The response also stated that although there was evidence of sulfidation found on the failed turbine, it is not linked directly to the cause of failure of the blades. In the absence of such a link, TC deemed it appropriate to continue to rely on visual inspection and performance runs as the means for limiting the effects of turbine blade sulfidation. Subsequent to TC's first response, the TSB Engineering Branch conducted a further examination and testing, and documented its findings in report LP037/04. On reading that report, TC provided a second response, dated 15April2004, in which it concluded that sulfidation was indeed the process by which the turbine blades deteriorated and eventually failed due to fatigue. TC will recommend to the engine manufacturer that an inspection for sulfidation on the turbine blades of this engine be implemented, and that this recommendation will be tracked in TC's Civil Aviation database.Safety Action Taken On 18 July 2003, the TSB sent a Safety Information Letter to TC, with copies to Bell Helicopter Textron; Aeronautical Accessories, Inc.; British Columbia Forest Service; and Cariboo Chilcotin Helicopters Ltd. The letter described the conditions that resulted in the pilot's door being blocked by the after-market step. This information was provided to TC, to take whatever action it deemed necessary. On 03 October 2003, the TSB sent a Safety Advisory (A030018) to TC, with copies to Rolls-Royce, Bell Helicopter Textron, and Cariboo Chilcotin Helicopters Ltd., drawing attention to the predominant cause for the engine power loss - sulfidation corrosion. Because periodic inspections are not required, the risk of this corrosion going undetected in other aircraft powered by Rolls-Royce 250engines remains. The letter stated that TC may wish to consider whether periodic inspections should be carried out to help detect damage caused by sulfidation corrosion. TC responded to the Safety Advisory on 05February2004, suggesting that in the absence of data substantiating the need for scheduled washings of the compressor (and turbine) blades, the introduction of such a schedule is deemed unnecessary. The response also stated that although there was evidence of sulfidation found on the failed turbine, it is not linked directly to the cause of failure of the blades. In the absence of such a link, TC deemed it appropriate to continue to rely on visual inspection and performance runs as the means for limiting the effects of turbine blade sulfidation. Subsequent to TC's first response, the TSB Engineering Branch conducted a further examination and testing, and documented its findings in report LP037/04. On reading that report, TC provided a second response, dated 15April2004, in which it concluded that sulfidation was indeed the process by which the turbine blades deteriorated and eventually failed due to fatigue. TC will recommend to the engine manufacturer that an inspection for sulfidation on the turbine blades of this engine be implemented, and that this recommendation will be tracked in TC's Civil Aviation database.